Research map / Video / Video-language understanding
In-Home Daily-Life Captioning Using Radio Signals
Lijie Fan, Tianhong Li, Yuan Yuan, Dina Katabi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use 3D skeletons from RF instead of person locations only (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 19.1 -> 26.7 | Improved |
| use 3D skeletons instead of 2D skeletons (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 22.3 -> 26.7 | Improved |
| add a floormap (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 22.1 -> 26.7 | Improved |
| use person-centric coordinate floormap instead of an image-based floormap (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 24.4 -> 26.7 | Improved |
| add noise to floormap measurements (20cm location, 10cm size, 30 degrees rotation) (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 26.7 -> 26.5 | No clear effect |
| add L2 feature alignment on paired data (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 20.3 -> 26.7 | Improved |
| add discriminator alignment on unpaired data (ablation) | RF-Diary · RCD test (visible scenario) | CIDEr: 24.6 -> 26.7 | Improved |
| add L2 feature alignment on paired data (ablation) | RF-Diary · RCD test (visible scenario) | BLEU-4: 18.5 -> 23.5 | Improved |
| use RF-Diary instead of HRL video captioning | RF-Diary vs HRL · RCD test (visible scenario) | CIDEr: 27.0 -> 26.7 | No clear effect |
| move from visible to dark/occlusion scenario | RF-Diary · RCD test (visible scenario) | CIDEr: 26.7 -> 25.9 | No clear effect |
Closest papers by meaning
- ParaHome: Parameterizing Everyday Home Activities Towards 3D Generative Modeling of Human-Object Interactions (CVPR 2025 · similarity 0.81)
- Video ReCap: Recursive Captioning of Hour-Long Videos (CVPR 2024 · similarity 0.80)
- ChatHuman: Chatting about 3D Humans with Tools (CVPR 2025 · similarity 0.79)
- Dense Video Object Captioning from Disjoint Supervision (ICLR 2025 · similarity 0.79)
- Image Captioning with Multi-Context Synthetic Data (AAAI 2024 · similarity 0.79)
- SensorLM: Learning the Language of Wearable Sensors (NeurIPS 2025 · similarity 0.79)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
